An adaptive fuzzy C-means algorithm for image segmentation in the presenceof intensity inhomogeneities

We present a novel algorithm for obtaining fuzzy segmentations of images th at are subject to multiplicative intensity inhomogeneities, such as magneti c resonance images. The algorithm is formulated by modifying the objective function in the fuzzy C-means algorithm to include a multiplier field, whic h allows the centroids for each class to vary across the image. First and s econd order regularization terms ensure that the multiplier field is both s lowly varying and smooth. An iterative algorithm that minimizes the objecti ve function is described, and its efficacy is demonstrated on several test images. (C) 1999 Elsevier Science B.V. All rights reserved.